mirror of
https://github.com/Termix-SSH/Termix.git
synced 2026-10-28 14:15:24 +00:00
ZacharyZcR
bf817d8bd1
* feat(ai): add host coding agent sessions with configured providers * Wait for agent initialization and shutdown before resuming sessions * Keep Pi transcripts outside its configuration migration directory * Cover Pi session directory separation * Update core permission catalog expectations for remote agents * Install remote agent runtimes from verified official sources * Wait for the identity panel credentials request in its test * Add scoped SSH forwarding setup for remote agents * Verify forwarding setup through a fresh SSH connection * Bound the SSH forwarding verification timeout * Return the forwarding verification callback result * fix(ai): support agent send and interrupt keyboard shortcuts * feat(ai): add agent queues attachments worktrees and session recovery * fix(ai): handle deleted workspace paths and duplicate status events * fix(ai): accept file references in root workspaces * fix(ai): refresh paused queues after agent errors * fix(ai): store agent sessions in plugin tables and lock down OpenCode Agent sessions lived in a single ctx.kv value holding up to 2000 events, which passed the 256 KB kv limit within minutes of a normal run. Every save after that failed, so history, nativeId, queue and archive changes were lost and every event wrote a kv audit row. Sessions, events and queued prompts now have their own tables. A running agent appends events in batches at most once a second with a lazy save; session and queue changes save immediately. Prompts, drafts and stored event text are capped in UTF-8 bytes to fit MySQL TEXT. kv:own is no longer requested. The OpenCode server now requires a per-session password, so other local accounts on the host cannot drive the agent through its loopback port. * test(upgrade): include agent session tables in the reviewed upgrade SQL
AI Assistant
An AI assistant that can read your hosts, snippets, fleets, alerts and automations, and suggest changes for you to approve.
Features
- Chat with an assistant that can read your hosts, snippets, fleets, alerts, automations, workspaces, network graph and homepage.
- The assistant suggests changes, and nothing is applied until you approve it.
- Can run read-only diagnostic commands on a host if you allow it.
- Open the assistant from the terminal toolbar with the current session as context.
- Works with Ollama, Anthropic, OpenAI, Google Gemini and any OpenAI compatible provider.
Settings
Admin
- AI assistant: let users turn the assistant on. While this is off it is hidden for everyone.
- Allowed private AI hosts: hosts on your private network a provider may point at, such as a self-hosted Ollama.
User
- Enable the AI assistant: turn the assistant on for yourself.
- Allow read-only diagnostic commands: let the assistant run safe commands on your hosts.
- Providers: the AI providers, models and API keys you use.
Host
- Enable AI Assistant: allow the assistant to work with this host.
Permissions
ai.use: Open the assistant and ask it things. Admins and users have it by default.ai.manage_providers: Add, edit and remove AI providers and their API keys. Admins and users have it by default.ai.apply_proposals: Let the assistant carry out the changes it suggests. Admins and users have it by default.ai.services.use: Let other features call the assistant for you. Admins and users have it by default.ai.secrets.share: Let other features use the stored AI API key without seeing it. Admins and users have it by default.
Services
Uses from other plugins:
workspaces.saved,network-topology.graph,snippets.access,fleets.access,terminal.history,automations.accessandhomepage.itemsto read your data. Each one is optional.
Development
npm run build # build into dist/
npm run test # run this plugin's tests
npm run typecheck # type-check this plugin